Projection position preset method and system

Abstract

The application discloses a kind of projection position preset method and system, including obtaining the digital fluoroscopy image of current projection position of patient, the preprocessed image is obtained to digital fluoroscopy image;The key feature and variation feature are obtained to the feature extraction of preprocessed image;The segmented image is obtained based on key feature and variation feature to digital fluoroscopy image segmentation identification, the segmented image is labeled, and the segmented image after labeling is obtained;Modeling is carried out based on segmented image after labeling, and the rotatable anatomical model is obtained, based on rotatable anatomical model, the key anatomical feature point is obtained, the key anatomical feature point is stored as calibration value, and structured report is generated based on rotatable anatomical model.The application can significantly improve the accuracy and efficiency of projection position preset in operation, realize the accurate adjustment and rapid reproduction of projection position, reduce the operation time and operation difficulty, reduce the operation risk, and improve the operation success rate.

Description

Technical Field

[0001] This invention belongs to the field of medical technology, specifically relating to a method and system for presetting the body position for imaging. Background Technology

[0002] In medical imaging examinations and surgical procedures, setting and adjusting the patient's position is a fundamental and crucial operational step. The imaging position refers to the specific posture, angle, and location that the patient needs to adopt in order to obtain clear and accurate medical images. With the development of modern medical technology, higher requirements have been placed on the precision and stability of the imaging position setting. Precise imaging position can significantly improve the accuracy of image diagnosis, ensure good exposure of the surgical field, reduce the risk of tissue damage, and improve surgical efficiency and safety.

[0003] In existing technologies, adjusting the patient's position mainly relies on the experience and judgment of medical staff and manual operation. However, this existing adjustment method has several inherent defects. For example, it is inefficient during manual adjustment, requiring repeated attempts and corrections. When fine adjustments are needed or the patient's cooperation is not high, it takes even longer, which may prolong the patient's examination or anesthesia time, leading to patient discomfort or even risks. Furthermore, since manual operation is highly dependent on the operator's experience level and condition, it is very easy to make adjustment errors due to various reasons, making it difficult to guarantee that the required position can be accurately reproduced every time.

[0004] Therefore, how to provide an effective technical solution to address the problems of low efficiency and insufficient accuracy in existing technologies, making it difficult to guarantee accurate reproduction of the required body position each time, has become an urgent problem to be solved in existing technologies. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for presetting the projection position, so as to solve the above-mentioned problems existing in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for presetting the imaging position, comprising: Acquire a digital fluoroscopic image of the patient in the current projection position, preprocess the digital fluoroscopic image, and obtain the preprocessed image; Feature extraction is performed on the preprocessed image to obtain key features and variant features; The digital perspective image is segmented and identified based on key features and variation features to obtain a segmented image. The segmented image is then labeled to obtain a labeled segmented image. Modeling is performed on the labeled segmented images to obtain a rotatable anatomical model. Based on the rotatable anatomical model, key anatomical feature points are obtained and stored as calibration values.

[0007] In one possible design, the digital perspective image is preprocessed to obtain a preprocessed image, including: An adaptive filtering algorithm is used to denoise the digital perspective image to obtain a denoised digital perspective image. Wavelet transform is used to process the motion artifacts in the denoised digital perspective image to obtain the processed digital perspective image. Histogram equalization is used to perform image enhancement on the processed digital perspective image to obtain the enhanced digital perspective image. Spatial alignment is performed on the enhanced digital perspective image to obtain the preprocessed image.

[0008] In one possible design, feature extraction is performed on the preprocessed image to obtain key features and variant features, including: The GRSDA algorithm is used to extract features from the preprocessed image to obtain global anatomical features and local microstructure features. Global anatomical features and local microstructural features are encoded to obtain global high-dimensional feature codes and local high-dimensional feature codes; Key features are obtained by dimensionality reduction of global and local high-dimensional feature codes using principal component analysis. The key features are processed using a deep autoencoder to obtain variant features.

[0009] In one possible design, the GRSDA algorithm is constructed based on genetic algorithms and rough set theory; feature extraction is performed on the preprocessed image based on the GRSDA algorithm to obtain global anatomical features and local microstructure features, including the following steps: Step 1: Extract the original features from the preprocessed image to obtain multiple sets of original image feature subsets. Encode each set of original image feature subsets into binary to obtain multiple image chromosomes. The image chromosomes are used to represent whether a feature is selected. Step 2: Construct a decision table based on the original feature subset of each image group. The rows of the decision table represent image regions, and the columns represent feature attributes and classification labels. Attribute dependencies are obtained based on rough set theory and the decision table. Based on the attribute dependencies, the fitness function is obtained. Step 3: Calculate the fitness of the image chromosome using the fitness function, select the image chromosome with a fitness greater than the preset value using the roulette wheel method, and take the image chromosome with a fitness greater than the preset value as the sub-chromosome. Randomly select one sub-chromosome and exchange the gene fragments of a random number of segments of the selected sub-chromosome to obtain the exchanged sub-chromosome. Randomly flip the genes in the exchanged sub-chromosome according to the preset probability to obtain the mutated image chromosome. Step four: Repeat steps two and three until the preset iteration termination condition is met to obtain the optimal feature subset, which includes global anatomical features and local microstructure features.

[0010] In one possible design, the fitness function is calculated as follows: ; In the above formula, For the fitness function, For attribute dependency, These are the weighting coefficients. The sparsity of the original feature subset of the image.

[0011] In one possible design, the digitized perspective image is segmented and identified based on key features and variation features to obtain a segmented image. The segmented image is then labeled to obtain a labeled segmented image, including: Using the U-Net deep learning model, coarse-grained organ segmentation is performed on the digital perspective image based on key features and variation features to obtain the initial segmentation image; The initial segmented image is optimized based on the active contour model to obtain a segmented image; Identify and segment the image, and perform semantic annotation on the segmented image to obtain the annotated segmented image.

[0012] In one possible design, after obtaining the segmented image, the following is also included: A heterogeneous classification model is constructed based on support vector machine and random forest models; Heterogeneous classification models are used to identify segmented images and obtain abnormal anatomical features; The abnormal anatomical features are weighted according to the attention mechanism to obtain the weighted segmentation image.

[0013] Secondly, the present invention provides a projection positioning preset system, including an external composite operating table. The composite operating table includes an angle sensor, a pull-wire sensor, a PID controller, and a servo motor. The angle sensor is used to collect the actual angles of the front-to-back rotation, left-to-right rotation, backboard rotation, left legboard rotation, and right legboard rotation. The pull-wire sensor is used to collect the lifting height of the composite operating table. The PID controller is used to send control commands to the servo motor. The servo motor is used to receive the control commands issued by the PID controller and control the composite operating table according to the control commands. The system also includes an imaging device, an anatomical analysis module, and a closed-loop control module. The anatomical analysis module is communicatively connected to the imaging device and the closed-loop control module, respectively. The imaging device is used to acquire images of the patient's current projection position and convert the images of the patient's current projection position into digital fluoroscopic images of the patient's current projection position. The anatomical analysis module is used to implement the projection position preset method described in the first aspect, and uploads the calibration value to the closed-loop control module; The closed-loop control module is communicatively connected to the imaging equipment and the hybrid operating table. The closed-loop control module is used to collect spatial coordinate data of the hybrid operating table and the imaging equipment. Based on the spatial coordinate data of the hybrid operating table and the imaging equipment, the anatomical points are obtained, and the difference between the anatomical points and the calibration values ​​is calculated. The module also acquires data collected by the angle sensor and the wire sensor, constructs a spatial position model of the operating table based on the data collected by the angle sensor and the wire sensor, and adjusts the imaging equipment and / or the hybrid operating table based on the difference between the anatomical points and the calibration values ​​and the spatial position model of the operating table.

[0014] In one possible design, the closed-loop control module is also used to establish the inverse equation of the six-degree-of-freedom parallel mechanism and the patient's respiratory motion prediction model, and adjust the torque output of the servo motor according to the inverse equation of the six-degree-of-freedom parallel mechanism and the patient's respiratory motion prediction model to adjust the composite operating table.

[0015] Thirdly, the present invention provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the projection position preset method as described in the first aspect.

[0016] The beneficial effects of this invention are as follows: This invention discloses a method and system for presetting the patient's imaging position. The method includes acquiring a digital fluoroscopic image of the patient's current imaging position; preprocessing the digital fluoroscopic image to obtain a preprocessed image; extracting features from the preprocessed image to obtain key features and variant features; segmenting and recognizing the digital fluoroscopic image based on the key features and variant features to obtain a segmented image; annotating the segmented image to obtain an annotated segmented image; modeling based on the annotated image to obtain a rotatable anatomical model; obtaining key anatomical feature points based on the rotatable anatomical model; storing the key anatomical feature points as calibration values; and generating a structured report based on the rotatable anatomical model. This invention adjusts the imaging equipment and / or the hybrid operating table by calculating the difference between the closed-loop control module and the calibration values ​​to achieve precise adjustment and rapid reproduction of the imaging position, reducing surgical time and operational difficulty. Simultaneously, precise imaging position reduces surgical risks, improves patient experience, reduces the number of intraoperative fluoroscopy sessions, reduces radiation exposure, increases surgical success rate, facilitates patient recovery, and is easy to apply and promote. Attached Figure Description

[0017] Figure 1 A flowchart of the projection position preset method provided in the first aspect of the embodiment; Figure 2 This is a block diagram of the projection position preset system provided in the second aspect of the embodiment. Detailed Implementation

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.

[0019] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.

[0020] Example: like Figure 1 As shown, the first aspect of this embodiment provides a method for presetting the imaging position, which can be executed, but is not limited to, by a computer device or virtual machine with certain computing resources, such as a personal computer or smartphone, or by a virtual machine; the method for presetting the imaging position includes, but is not limited to, the following steps: S1. Acquire a digital fluoroscopic image of the patient in the current projection position, preprocess the digital fluoroscopic image, and obtain the preprocessed image; Specifically, in step S1, the digitized perspective image is preprocessed to obtain a preprocessed image, including: S11. An adaptive filtering algorithm is used to denoise the digital perspective image to obtain a denoised digital perspective image. S12. Use wavelet transform to process the motion artifacts in the denoised digital perspective image to obtain the processed digital perspective image; S13. Use histogram equalization to perform image enhancement on the processed digital perspective image to obtain the enhanced digital perspective image; Specifically, histogram equalization is an image processing technique that enhances contrast by adjusting the gray-level distribution of an image. Its principle is to transform the gray-level probability density distribution of the original image into a uniform distribution, thereby expanding the dynamic range of pixels and improving the clarity of image details. Image enhancement operations are performed on the processed digital perspective image to enhance the image contrast of the processed digital perspective image and highlight edge and texture details.

[0021] S14. Perform spatial alignment on the enhanced digital perspective image to obtain the preprocessed image.

[0022] Specifically, feature points of the enhanced digital perspective image can be extracted using SIFT (Scale-Invariant Feature Transform) or SURF (Speeded Up Robust Features) algorithms. These feature points are then matched to achieve spatial alignment of the image, ensuring the coordinate consistency of the anatomical structures. Both SIFT and SURF algorithms are existing technologies and will not be elaborated upon here.

[0023] Furthermore, the mutual information maximization algorithm can be used to optimize the alignment accuracy. The principle of the mutual information maximization algorithm is to optimize the model parameters to maximize the mutual information between the input and output, thereby enhancing the model performance.

[0024] S2. Perform feature extraction on the preprocessed image to obtain key features and variant features; Specifically, in step S2, feature extraction is performed on the preprocessed image to obtain key features and variant features, including: S21. Based on the GRSDA algorithm, feature extraction is performed on the preprocessed image to obtain global anatomical features and local microstructure features; Among them, the GRSDA algorithm is based on genetic algorithm and rough set theory. Genetic algorithm can be used to optimize the feature selection process by iteratively screening the optimal feature subset. Rough set theory can handle the uncertainty in medical images by extracting key features through attribute reduction. The attribute reduction of rough set theory can eliminate honor features, and the mutation operation of genetic algorithm can enhance the robustness to noise.

[0025] Specifically, in step S21, feature extraction is performed on the preprocessed image based on the GRSDA algorithm to obtain global anatomical features and local microstructure features, including the following steps: Step 1: Extract the original features from the preprocessed image to obtain multiple sets of original image feature subsets. Encode each set of original image feature subsets into binary to obtain multiple image chromosomes. The image chromosomes are used to represent whether a feature is selected. The original image features in the original image feature subset include, but are not limited to, shape moments, HOG texture, and wavelet coefficients.

[0026] Step 2: Construct a decision table based on the original feature subset of each image group. The rows of the decision table represent image regions, and the columns represent feature attributes and classification labels. Attribute dependencies are obtained based on rough set theory and the decision table. Based on the attribute dependencies, the fitness function is obtained. The classification labels indicate whether the condition is normal or pathological.

[0027] Step 3: Calculate the fitness of the image chromosome using the fitness function, select the image chromosome with a fitness greater than the preset value using the roulette wheel method, and take the image chromosome with a fitness greater than the preset value as the sub-chromosome. Randomly select one sub-chromosome and exchange the gene fragments of a random number of segments of the selected sub-chromosome to obtain the exchanged sub-chromosome. Randomly flip the genes in the exchanged sub-chromosome according to the preset probability to obtain the mutated image chromosome. Step four: Repeat steps two and three until the preset iteration termination condition is met to obtain the optimal feature subset, which includes global anatomical features and local microstructure features.

[0028] The preset iteration termination condition can be set to 50 iterations or fitness convergence; global anatomical features include, but are not limited to, organ shape, and local microstructural features include, but are not limited to, vascular bifurcation points and trabecular texture; global anatomical features and local microstructural features are dynamically weighted through attribute dependency to avoid the bias of manually set weights.

[0029] The expression for calculating attribute dependency in step two is as follows: ; In the above formula, For attribute dependency, The positive classification region is obtained based on a subset of the original features of the current image. For the total number of image regions, when A higher value indicates a greater contribution of the original subset of image features to the classification.

[0030] The expression for calculating the fitness function in step two is as follows: ; In the above formula, For the fitness function, For attribute dependency, These are the weighting coefficients. The sparsity of the original feature subset of the image. Used to balance classification performance and feature reduction.

[0031] S22. Encode global anatomical features and local microstructural features to obtain global high-dimensional feature codes and local high-dimensional feature codes; S23. Principal component analysis is used to reduce the dimensionality of global and local high-dimensional feature codes to obtain key features; S24. Process key features based on deep autoencoders to obtain variant features.

[0032] S3. Segment and identify the digital perspective image based on key features and variation features to obtain a segmented image. Then, annotate the segmented image to obtain an annotated segmented image. Specifically, in step S3, the digitized perspective image is segmented and identified based on key features and variation features to obtain a segmented image. The segmented image is then labeled to obtain a labeled segmented image, including: S31. Using the U-Net deep learning model, coarse-grained organ segmentation is performed on the digitized perspective image based on key features and variation features to obtain the initial segmentation image; S32. Optimize the initial segmented image based on the active contour model to obtain a segmented image; The ActiveContour model works by evolving an initial curve by minimizing a defined energy function. This energy function typically consists of two parts: internal energy (which keeps the curve smooth) and external energy (which attracts the curve towards the target boundary). Through iterative optimization, the curve gradually converges to the minimum energy value, thus accurately depicting the contour of the target object and solving the problem of separating adherent tissues.

[0033] S33. Identify the segmented image and perform semantic annotation on the segmented image to obtain the annotated segmented image.

[0034] This includes semantic annotation of segmented images, as well as the construction of knowledge graphs, which automatically associate anatomical terms, such as the MeSH terminology system, and support multilingual label mapping to avoid the limitations of a single language.

[0035] Furthermore, after obtaining the segmented image, the process also includes: A heterogeneous classification model is constructed based on support vector machine and random forest models; Heterogeneous classification models are used to identify segmented images and obtain abnormal anatomical features; The abnormal anatomical features are weighted according to the attention mechanism to obtain the weighted segmentation image.

[0036] In one possible design, an LSTM network is used to analyze the patterns of anatomical structural deformation in the weighted segmented image and generate a probability map to predict the development path of lesions and assist doctors in making treatment decisions.

[0037] S4. Model the labeled segmented image to obtain a rotatable anatomical model. Based on the rotatable anatomical model, obtain key anatomical feature points and store them as calibration values.

[0038] Specifically, the Moving Cubes (MC) algorithm is used to generate a rotatable anatomical model to support observation of any cross-section. The Moving Cubes algorithm is an existing technology and will not be described in detail here. At the same time, AR technology is integrated to realize real-time projection positioning during surgery and generate a structured report based on the rotatable anatomical model.

[0039] This embodiment provides a method for pre-setting the imaging position during surgery, which can significantly improve the accuracy and efficiency of pre-setting the imaging position during surgery, realize precise adjustment and rapid reproduction of the imaging position, reduce surgical time and operational difficulty, and at the same time help doctors to more accurately judge the surgical situation, reduce surgical risks, reduce the number of intraoperative fluoroscopy sessions and radiation exposure, improve the success rate of surgery, and benefit patient recovery.

[0040] like Figure 2 As shown, the second aspect of this embodiment provides a projection position preset system, including an external composite operating table. The composite operating table includes an angle sensor, a pull-wire sensor, a PID controller, and a servo motor. The angle sensor is used to collect the actual angles of the front-to-back rotation, left-to-right rotation, backboard rotation, left legboard rotation, and right legboard rotation. The pull-wire sensor is used to collect the lifting height of the composite operating table. The PID controller is used to send control commands to the servo motor. The servo motor is used to receive the control commands issued by the PID controller and control the composite operating table according to the control commands. The system also includes an imaging device, an anatomical analysis module, and a closed-loop control module. The anatomical analysis module is communicatively connected to the imaging device and the closed-loop control module, respectively. The imaging device is used to acquire images of the patient's current projection position and convert the images of the patient's current projection position into digital fluoroscopic images of the patient's current projection position. The anatomical analysis module is used to implement the projection position preset method described in the first aspect, and uploads the calibration value to the closed-loop control module; The closed-loop control module is communicatively connected to the imaging equipment and the hybrid operating table. The closed-loop control module is used to collect spatial coordinate data of the hybrid operating table and the imaging equipment, obtain anatomical points based on the spatial coordinate data of the hybrid operating table and the imaging equipment, calculate the difference between the anatomical points and the calibration values, acquire data collected by angle sensors and wire sensors, construct a spatial position model of the operating table based on the data collected by angle sensors and wire sensors, and adjust the imaging equipment and / or the hybrid operating table based on the difference between the anatomical points and the calibration values ​​and the spatial position model of the operating table.

[0041] Preferably, the imaging equipment can be a C / G type arm.

[0042] In one possible design, the closed-loop control module is also used to establish the inverse equation of the six-degree-of-freedom parallel mechanism and the patient's respiratory motion prediction model, and adjust the torque output of the servo motor according to the inverse equation of the six-degree-of-freedom parallel mechanism and the patient's respiratory motion prediction model to adjust the composite operating table.

[0043] Among them, the six degrees of freedom refers to the fact that the composite operating table has a 6-degree-of-freedom electric control structure: bed panel lifting, bed front and back rotation, bed surface left and right rotation, back panel rotation, left leg panel rotation, and right leg panel rotation. The movement of each motor is controlled by a control model to reduce the difference between the actual coordinates and the calibration values.

[0044] The third aspect of this embodiment provides a computer program product, including a computer program or instructions, which, when executed by a computer, are used to implement the projection position preset method as described in the first aspect.

[0045] The working process, working details and technical effects of the aforementioned computer program product provided in this embodiment can be found in the projection position preset method as described in the first aspect, and will not be repeated here.

[0046] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for presetting the projection position, characterized in that, include: Acquire a digital fluoroscopic image of the patient in the current projection position, preprocess the digital fluoroscopic image, and obtain the preprocessed image; Feature extraction is performed on the preprocessed image to obtain key features and variant features; The digital perspective image is segmented and identified based on key features and variation features to obtain a segmented image. The segmented image is then labeled to obtain a labeled segmented image. Modeling is performed on the labeled segmented images to obtain a rotatable anatomical model. Based on the rotatable anatomical model, key anatomical feature points are obtained and stored as calibration values.

2. The method for presetting the projection position according to claim 1, characterized in that, The digital perspective image is preprocessed to obtain the preprocessed image, including: An adaptive filtering algorithm is used to denoise the digital perspective image to obtain a denoised digital perspective image. Wavelet transform is used to process the motion artifacts in the denoised digital perspective image to obtain the processed digital perspective image. Histogram equalization is used to perform image enhancement on the processed digital perspective image to obtain the enhanced digital perspective image. Spatial alignment is performed on the enhanced digital perspective image to obtain the preprocessed image.

3. The method for presetting the projection position according to claim 1, characterized in that, Feature extraction is performed on the preprocessed image to obtain key features and variant features, including: The GRSDA algorithm is used to extract features from the preprocessed image to obtain global anatomical features and local microstructure features. Global anatomical features and local microstructural features are encoded to obtain global high-dimensional feature codes and local high-dimensional feature codes; Key features are obtained by dimensionality reduction of global and local high-dimensional feature codes using principal component analysis. The key features are processed using a deep autoencoder to obtain variant features.

4. The method for presetting the projection position according to claim 3, characterized in that, The GRSDA algorithm is constructed based on genetic algorithms and rough set theory. Based on the GRSDA algorithm, feature extraction is performed on the preprocessed image to obtain global anatomical features and local microstructure features, including the following steps: Step 1: Extract the original features from the preprocessed image to obtain multiple sets of original image feature subsets. Encode each set of original image feature subsets into binary to obtain multiple image chromosomes. The image chromosomes are used to represent whether a feature is selected. Step 2: Construct a decision table based on the original feature subset of each image group. The rows of the decision table represent image regions, and the columns represent feature attributes and classification labels. Attribute dependencies are obtained based on rough set theory and the decision table. Based on the attribute dependencies, the fitness function is obtained. Step 3: Calculate the fitness of the image chromosome using the fitness function, select the image chromosome with a fitness greater than the preset value using the roulette wheel method, and take the image chromosome with a fitness greater than the preset value as the sub-chromosome. Randomly select one sub-chromosome and exchange the gene fragments of a random number of segments of the selected sub-chromosome to obtain the exchanged sub-chromosome. Randomly flip the genes in the exchanged sub-chromosome according to the preset probability to obtain the mutated image chromosome. Step four: Repeat steps two and three until the preset iteration termination condition is met to obtain the optimal feature subset, which includes global anatomical features and local microstructure features.

5. The method for presetting the projection position according to claim 4, characterized in that, The fitness function is calculated as follows: ； In the above formula, For the fitness function, For attribute dependency, These are the weighting coefficients. The sparsity of the original feature subset of the image.

6. The method for presetting the projection position according to claim 1, characterized in that, Segmentation and recognition of digitized perspective images are performed based on key features and variation features to obtain segmented images. These segmented images are then labeled to obtain labeled segmented images, including: Using the U-Net deep learning model, coarse-grained organ segmentation is performed on the digital perspective image based on key features and variation features to obtain the initial segmentation image; The initial segmented image is optimized based on the active contour model to obtain a segmented image; Identify and segment the image, and perform semantic annotation on the segmented image to obtain the annotated segmented image.

7. The method for presetting the projection position according to claim 1, characterized in that, After obtaining the segmented image, the following steps are also included: A heterogeneous classification model is constructed based on support vector machine and random forest models; Heterogeneous classification models are used to identify segmented images and obtain abnormal anatomical features; The abnormal anatomical features are weighted according to the attention mechanism to obtain a weighted segmentation image.

8. A projection positioning preset system, comprising an externally connected composite operating table, the composite operating table including an angle sensor, a pull-wire sensor, a PID controller, and a servo motor, wherein the angle sensor is used to acquire the actual angles of the table's front-to-back rotation, left-to-right rotation, backrest rotation, left legboard rotation, and right legboard rotation; the pull-wire sensor is used to acquire the composite operating table's lifting height; the PID controller is used to send control commands to the servo motor; and the servo motor is used to receive the control commands from the PID controller and control the composite operating table according to the control commands, characterized in that... It also includes imaging equipment, an anatomical analysis module, and a closed-loop control module, wherein the anatomical analysis module is communicatively connected to the imaging equipment and the closed-loop control module, respectively; The imaging device is used to acquire images of the patient's current projection position and convert the images of the patient's current projection position into digital fluoroscopic images of the patient's current projection position. The anatomical analysis module is used to implement the projection position preset method according to any one of claims 1-7, and uploads the calibration value to the closed-loop control module; The closed-loop control module is communicatively connected to the imaging equipment and the hybrid operating table. The closed-loop control module is used to collect spatial coordinate data of the hybrid operating table and the imaging equipment. Based on the spatial coordinate data of the hybrid operating table and the imaging equipment, the anatomical points are obtained, and the difference between the anatomical points and the calibration values ​​is calculated. The module also acquires data collected by the angle sensor and the wire sensor, constructs a spatial position model of the operating table based on the data collected by the angle sensor and the wire sensor, and adjusts the imaging equipment and / or the hybrid operating table based on the difference between the anatomical points and the calibration values ​​and the spatial position model of the operating table.

9. The imaging positioning system according to claim 8, characterized in that, The closed-loop control module is also used to establish the inverse equation of the six-degree-of-freedom parallel mechanism and the patient's respiratory motion prediction model, and adjust the torque output of the servo motor according to the inverse equation of the six-degree-of-freedom parallel mechanism and the patient's respiratory motion prediction model to adjust the composite operating table.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or the instructions are executed by the computer, they implement the projection position preset method as described in any one of claims 1 to 7.

Images